Dr. Pearson's Lectures

Object Recognition

The Process of matching a Representation of a viewed object to a representation in stored memory

Object Constancy

How we disregard variations of the representation to abstract the identity of the object

Template Matching

Perceived stimuli are compared to Templates stored in memory, if the templates match, the stimulus is recognized.

David Marr

Conceptualized vision as a problem of information processing

Attempts to define: How shape is represented in a way that is useful for recognition of 3D objects,

How he solved it: 1) Primal Sketch 2) 2.5 D Sketch 3) 3D model representation

Primal Sketch

A primitive but rich descrption of the way in which light information changes across the visual field
Captured the important underlying structure
Not ever luminance change is an edge
Nor all edges result in Luminance changes

2.5 D Sketch

Concerned with the recovery of surfaces and depth Information
From the POV of the observer
Doesn’t generalize to other viewpoints

3D Model Representation

Concerned with the recovery of volumetric parts that make up the object
Conceptualized a “GENERAL CONES”, hierarchically organized along principal axes of the object
Independent of the POV of the Observer

Irving Biederman

Developed the Recognition by components theory
All objects are made up of volumetric components (Geons)
36 Geons are sufficent to describe all objects

Geon Structural Descriptions

Objects are represented by sepcific geons and the relations between them

Edge Extraction

Parsing Regions of Concavity

People are sensitive to regions of concavity and spend more time looking at these regions
This is where geons connect

Detection of non-accidental properties

Properties that are tolerant to change in viewpoint, occlusion, noise, etc
ex: Straight v Curved lines, Parallel v Converging lines)

Determination of Components

Matching components to object representations

The Priming effect cannot be based on low level image features ( The specific edges and vertices) as these differed in the complementary images
Must be either: Geons or Full object representation

The Issue of Viewpoint

Objects are recognized on the basis of stored views that we have encountered through experiences, New views have to be matched to stored views

Mental Rotation

The act of imagining something turning around through space
Usually measured by asking people to discriminate between objects and their mirror images that are presented in different orientations

Viewpoint-dependent recognition

Observers take longer to name rotated objects
Reaction time is proportional to degree of misorientation
Effects of viewpoint decreases with familiarity of the rotated views

Harris,Harris,& Caine (2002)

Patient CB with damage to right Basal Ganglia following a stroke
Had imapirmmet in tasks requiring mental rotation
Intact recognition of rotated objects and sihouetts
Further evidence that mental rotation does not contribute to recognition of misoriented objects

“Views” are holistic representations

Parts and their relationships are not able to be recovered independently

Harris, Harris, & Caine (2001)

DeCaro and Reeves(2000)

Presented backward-masked rotated objects, to see how long people nee

Faces

3 types types of Face Recognition

Composite Effect, Part-whole Effect, Inversion effect

Prosopagnosia

Deficits in face recognition (With intact object recognition)

Visual O

Face Processing involves which areas of the brain?

Fusiform Face Area Occipital Face Area, and Superior Temporal Sulcus

Object Processing involved which areas of the brain?

Lateral and Inferior Occipital lobes , Inferior Temporal Love ( including the Fusiform Gyrus)

Face Pareidolia

Things that have a face configuration are seen as faces

First-order relations

Eyes above nose, nose above mouth

Holistic Processing

All features are percieved togetehr as a unified gestalt

Second-order relations

Exact Metric Distances between features

Composite Effect

It is difficult to judge weather the top part of a face is the same or different when presented with different halves
- its much easier when the parts of misaligned or inverted

Part-Whole Effect

It is easier to recognize a face feature when it is presented within a face

Face Inversion Effect

Upside down faces are harder to recognise
Parts are hard to discriminate in inverted faces
Recognition of whole faces is easier than recognition of individual features
Feature Recognition is equivalently bad when upright and inverted

Tanaka and Farah (1993)

Subjects learned intact/scrambled faces, then they had to identfy the individuals or isolated features

Schmalz, Palermo, Harris, and Coltheart (2009)

Young man with long-standing developmental prosopagnosia, had parriular difficulty discriminating features when they were presented in upright faces, BUT performed better for isolated features and inverted faces

Large Configural Changes leave Recognition Unharmed

Within-person variation is as big, if not bigger than between-person variation

Non-confgural changes harm recognition

Hard to recognize faces from photographic negatives
Drawings traced from photograpghs are quite hard to recognize but easier when texture is added

Expertiese Hypothesis

Face perception is an example of Visual Expertiese in individuating a visually homogenous class of objects
Expertiese leads to a cgange in the entry level of object recognition
Expertiese leads to holistic coding of objects of expertiese
Objects of expertiese recruit the FFA and OFA
There is interfearence bwteen stimuli tapping into shared exterteise processes

Episodic memory

Contains contexutal information about where and when some event took place

Semantic memory

Contains Facts and General information, and is generally free of experiental context

Autonetic Consciousness

Related to Episodic memory

Noetic Consciousness

Related to Semantic memory

Free Recall

Just spitting out all the information you remember

Serial Recall

Remembering a number sequence or a list of items

Cued Recall

Given a hint to help start the recall

Retrieval

the process of accessing a Target memory based on 1 or more cues and bringing it to awareness

What is being retrieved?

Retrieval leads to cortical reinstatement of pattern of activity present at encoding
Progression from a cue to a taregt memoey is through spreading activation of associative connections

What is the outcome when We are distracted while learning cues?

Distraction impairs retreival, especially if the distracting information is similalr

Relevance of Cues

Often search memory with inappropriate cues
We remember what we experience

Encoding-Specificity Principle

the Retrieval cue should have been present at the encoding

Cue-target Strength

Increased lef pre-frontal cortex activation for weakly associated cue-target pairs
Weak Association between target and cue requires more sustained attention and cognitive control

Number of Cues

The more cues, The better
Multiple cues often produced Super-additive effects

Target Strength

More effectivelty encoded memories are retrieved more easily

Retrieval Mode

Need to be in the right Cognitive set for a cue to trigger retrieval memory

Herron & Wilding (2004)

Study phase: Moving or not-moving or pleasantness, Left side or Right side
TEST PHASE: Cue before each trial
Operations: Was this word in the pleasantness task or motion task
Location: Was this word on the left or right?
Semantic: Moving or non moving
EEG traces and scalp topography during presentation of different cues
Retrieval mode engages the right frontal lobe

Context Cues

Spatial-Temporal
Mood
Psysiological (Caffience)
Cognitive

memory and time

More distant memories are more likely to be forgotten

Linton (1975)

Recorded daily events and randomly recalled some of them over a period of time
65% of items never retrieved were forgotten after 4 years
12% of those recalled 4x or more were forgotten after 4 years
Re-activating memories leads to the formation of Multiple Traces and re-consolication of more robust memories

Decay of the Memory Trace

Gradual Weakening of Connections between memory elements over time
Loss of Synaptic connections
Neurogenesis

Neurogenesis

The generation of new Neurons in the Hippocampus

Contextual Shift

A Mismatch between encoding and retrieval
Could also be explained by passage of time- content changes as time passses

Interference

Items associated with a Cue complete for access to awareness

Retroactive Interference

Forgetting caused by encoding new memory traces (especially similar ones)

Proactive Interference

Previously learned material prevents retrieval of more recent information

Retrieval-Induced Forgetting (RIF)

Anderson, Bjork & Bjork 1994

Participants learned 2 lists based on different Categories
Practice of Retrieval of a subset of one list (ex: Fruit A__)
Test recall of all items from both lists

Associative Competition

You practice Fruit-Apple more than Fruit-Banana and more readily retrieve Apple than Banana

Inhibiton

You may think of Banana, but you know that the fruit starts with A, so Banana is not important to you

Cue Independence

Retrieval of unpracticed items (Rp-) is poor even when they are cues with an unrelated associate

Retrieval Specificity

RIF is induced by retrieval practice but not simple re-study
Attempt to resolve competition at retrieval

Attention Dependence

RIF is reduced when retrieval practice is done under divided attention or in people with Attention Deficits

Interference Dependence

RIF is greater for higher- frequency item

Strength Independence

The degree if strengthening the practiced competitors (Rp+) does not preditc the size of RIF

Parts of the Brain associated with RIF

Ventral-Lateral prefrontal cortex (VLPFC) and Anterior Cingulate Cortex (ACC)

Directed Forgetting

A form of motivated forgetting that prevents elaborative encoding and rehersal

List method

Affects retrieval but not the recognition of items

Retrieval Inhibition Hypothesis

Instructions to forget inhibit retrieval of list one items

Context Shift Hypothesis

Instructions to forget separated list one items from List 2 context

Retrieval Supression

A form of motivated forgetting that prevents retireval in response to cues

Thought Substitution

An alertentaive way to avoid retrieving the target memory